Gesture Fingerprinting for Ad Visibility Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ad fraud, particularly ad fraud involving bot traffic and misrepresentation of ad views, leads to advertisers paying for views that were not actually visible to users, as existing technologies fail to distinguish between human and bot inputs effectively.
Innovation Solution
A computer-implemented method and system for detecting visibility of an application using gesture fingerprinting technology, which captures sensor data such as accelerometer data and pixel information to determine if an advertisement was viewed by comparing the data to predefined criteria, thereby distinguishing between human and bot interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional view counting methods are used to track advertisement views, then the system is simple to implement, but the reliability of view data is compromised due to inability to distinguish human from bot traffic
Solution Approach 1:
The patent segments the view detection process into multiple independent components: gesture capture module, fingerprint generation module, comparison module, and visibility determination module. Each component handles a specific aspect of the detection process, allowing the system to achieve high reliability through multi-factor analysis while maintaining manageable complexity through modular architecture.
Solution Approach 2:
The patent introduces gesture fingerprints as an intermediary layer between raw sensor data and view validation decisions. The fingerprint generation module transforms complex sensor data into comparable signatures, and the comparison module uses these fingerprints to determine visibility, acting as a mediator that simplifies the decision-making process while improving reliability.
2Reliability
If gesture fingerprinting with sensor data capture is implemented to detect visibility, then the reliability of distinguishing human from bot interactions is improved, but the use of energy and device complexity increase
Solution Approach 1:
The patent implements partial action by capturing sensor data only during specific gesture events rather than continuously. The system activates the sensor capture and fingerprint generation only when gesture inputs are detected, reducing overall energy consumption while maintaining sufficient data for reliable human-bot distinction during actual interaction periods.
Solution Approach 2:
The patent changes the operational parameters of the sensor system by adjusting sampling rates and activation thresholds based on device state and interaction context. The system dynamically modifies when and how sensor data is captured, reducing energy consumption during idle periods while ensuring adequate data collection during active viewing periods to maintain reliability.
Data Source
AI summary
Various implementations related to visibility detection using gesture fingerprinting are described. In one such implementation, a computer-implemented method includes receiving an input triggering a visible application to be presented on a display device, capturing visibility data associated with the visible application and the input, determining a visibility of the visible application by comparing the visibility data with visibility criteria, and storing the determined visibility of the visible application.